Kimi K3 vs Kimi K3.1 Pro
Kimi K3.1 Pro leads llmrun Score 142.8 to 132.7; neither fits 24 GB.
Run Kimi K3.1 Pro in the cloud →
Model A · Moonshot AI
Kimi K32779.9B · 105.8B active · 1M context · Other · Jun 2026
Model B · Moonshot AI
Kimi K3.1 Pro2780.1B · 106B active · 1M context · Other · Nov 2026
Choose Kimi K3.1 Pro if you want…
Higher llmrun Score
142.8 vs 132.7
Stronger at coding
140.7 vs 134.5
Ahead on the benchmarks
4 of 4 shared boards
Newer release
Nov 2026 vs Jun 2026
Kimi K3: no clear edge over Kimi K3.1 Pro in the data we have.
On your hardware
Each model at its default quant (Q4_K_M), a smaller one when it must. Estimates, ±20%.
8 GB
RTX 4060 · 8 GB tier
Kimi K3
Doesn't fit
smallest quant needs 766.5 GB
Kimi K3.1 Pro
Doesn't fit
smallest quant needs 1225.5 GB
12 GB
RTX 3060 12GB · 12 GB tier
Kimi K3
Doesn't fit
smallest quant needs 766.5 GB
Kimi K3.1 Pro
Doesn't fit
smallest quant needs 1225.5 GB
16 GB
RTX 5060 Ti 16GB · 16 GB tier
Kimi K3
Doesn't fit
smallest quant needs 766.5 GB
Kimi K3.1 Pro
Doesn't fit
smallest quant needs 1225.5 GB
24 GBreference
RTX 4090 · 24 GB tier
Kimi K3
Doesn't fit
smallest quant needs 766.5 GB
Kimi K3.1 Pro
Doesn't fit
smallest quant needs 1225.5 GB
32 GB
RTX 5090 · 32 GB tier
Kimi K3
Doesn't fit
smallest quant needs 766.5 GB
Kimi K3.1 Pro
Doesn't fit
smallest quant needs 1225.5 GB
48 GB
RTX A6000 · 48 GB tier
Kimi K3
Doesn't fit
smallest quant needs 766.5 GB
Kimi K3.1 Pro
Doesn't fit
smallest quant needs 1225.5 GB
Mac 32 GB
Mac Mini M4 · Macs
Kimi K3
Doesn't fit
smallest quant needs 766.5 GB
Kimi K3.1 Pro
Doesn't fit
smallest quant needs 1225.5 GB
Mac 64 GB
Mac Studio M4 Max · Macs
Kimi K3
Doesn't fit
smallest quant needs 766.5 GB
Kimi K3.1 Pro
Doesn't fit
smallest quant needs 1225.5 GB
Mac 128 GB
Mac Studio M4 Max · Macs
Kimi K3
Doesn't fit
smallest quant needs 766.5 GB
Kimi K3.1 Pro
Doesn't fit
smallest quant needs 1225.5 GB
Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).
Overall
Kimi K3
Kimi K3.1 Pro
Coding
Kimi K3
Kimi K3.1 Pro
Agents
Kimi K3
Kimi K3.1 Pro
no data
Math
Kimi K3
Kimi K3.1 Pro
no data
Science
Kimi K3
Kimi K3.1 Pro
Reasoning
Kimi K3
Kimi K3.1 Pro
no data
Kimi K3 · #4 of 106 open models · #4 of 106 that fit ≤2 TB
Kimi K3.1 Pro · #1 of 106 open models · #1 of 106 that fit ≤2 TB
Kimi K3
59.5%#3/72 open
Lab-reportedEpoch AI · Jul 2026
Kimi K3.1 Pro
60.8%#2/72 open
Lab-reportedEpoch AI · Oct 2026
Kimi K3
1524.5#4/48 open
Lab-reportedEpoch AI · Jul 2026
Kimi K3.1 Pro
1932.8#1/48 open
Lab-reportedEpoch AI · Oct 2026
Kimi K3
23.4%#4/84 open
Lab-reportedEpoch AI · Jul 2026
Kimi K3.1 Pro
24.6%#2/84 open
Lab-reportedEpoch AI · Oct 2026
Kimi K3
87.0%#2/31 open
Lab-reportedEpoch AI · Jul 2026
Kimi K3.1 Pro
91.0%#1/31 open
Lab-reportedEpoch AI · Oct 2026
36 more boards (other shared and single-model boards)
Only Kimi K3 (36)
- GPQA Diamond93.1%
- LMArena Text1488.2
- LiveBench Coding81.5
- LiveBench Reasoning90.7
- AIME 2024/202597.2%
- FrontierMath72.2%
- LMArena WebDev1656.4
- LiveBench Math84.4
- ARC-AGI-260.4%
- ARC-AGI94.5%
- SimpleQA50.6%
- SimpleBench60.7%
- FrontierMath Tier 439.0%
- WeirdML82.6%
- DeepSWE68.5%
- FrontierCode44.2%
- APEX-Agents47.8%
- Chess Puzzles37.0%
- DTBench91.2%
- ForecastBench64.7
- Furniture Assembly34.2%
- LiveBench 2026 Reasoning90.7
- LMCA52.7%
- Mystery Game Puzzles26.0%
- Vending-Bench 25165.0
- GDP.pdf19.0%
- LiveBench 2026 Browser16.0
- LiveBench 2026 Data Analysis78.7
- FrontierSWE25.9%
- LiveBench 2026 Coding59.3
- LiveBench 2026 Long-Horizon42.8
- Surface Evolver Bench95.0%
- GBA Eval48.4%
- LiveBench 2026 Design77.2
- WeirdML v37.5%
- LiveBench 2026 Research79.0
VRAM by quantization
Memory to load the model and 2K context; below it, the total at 32K and 128K context. Lighter side in bold.
Q2_K
Kimi K3
1187 GB928.6 GB file
32K 1185 GB · 128K 1196 GB
Kimi K3.1 Pro
1187 GB~1181.5 GB file
32K 1269 GB · 128K 1531 GB
Q3_K_M
Kimi K3
1361 GB~1355.2 GB file
32K 1359 GB · 128K 1370 GB
Kimi K3.1 Pro
1361 GB~1355.3 GB file
32K 1443 GB · 128K 1705 GB
Q4_K_M
Kimi K3
1674 GB~1668.0 GB file
32K 1672 GB · 128K 1682 GB
Kimi K3.1 Pro
1674 GB~1668.1 GB file
32K 1756 GB · 128K 2018 GB
Q5_K_M
Kimi K3
1986 GB~1980.7 GB file
32K 1985 GB · 128K 1995 GB
Kimi K3.1 Pro
1987 GB~1980.8 GB file
32K 2069 GB · 128K 2331 GB
Q6_K
Kimi K3
2299 GB~2293.4 GB file
32K 2297 GB · 128K 2308 GB
Kimi K3.1 Pro
2299 GB~2293.6 GB file
32K 2381 GB · 128K 2643 GB
Q8_0
Kimi K3
2786 GB~2779.9 GB file
32K 2784 GB · 128K 2794 GB
Kimi K3.1 Pro
2786 GB~2780.1 GB file
32K 2868 GB · 128K 3130 GB
2 more quants
IQ2_XXS
Kimi K3
770 GB711.1 GB file
32K 768 GB · 128K 779 GB
Kimi K3.1 Pro
not listed
BF16
Kimi K3
5566 GB~5559.9 GB file
32K 5564 GB · 128K 5574 GB
Kimi K3.1 Pro
5566 GB~5560.2 GB file
32K 5648 GB · 128K 5910 GB
Kimi K3
Ollama · Q4_K_M
ollama run hf.co/unsloth/Kimi-K3-GGUFKimi K3.1 Pro
No local recipe · Q4_K_M
No GGUF repo entry exists for this model yet.
All ways to run Kimi K3.1 Pro →| Spec | Kimi K3 | Kimi K3.1 Pro |
|---|---|---|
| Publisher | Moonshot AI | Moonshot AI |
| Parameters | 2779.9B | 2780.1B |
| Active parameters | 106B per token | 106B per token |
| Architecture | KimiK3ForConditionalGeneration | KimiK31ForConditionalGeneration |
| Context | 1M tokens | 1M tokens |
| Modalities | Vision, Chat | Vision, Chat |
| License | Other | Other |
| Released | Jun 2026 | Nov 2026 |
Frequently Asked Questions
Which is better, Kimi K3 or Kimi K3.1 Pro?
Kimi K3.1 Pro leads llmrun Score 142.8 to 132.7; neither fits 24 GB. Choose Kimi K3.1 Pro for higher llmrun Score and stronger at coding.
Which needs less VRAM, Kimi K3 or Kimi K3.1 Pro?
At Q4_K_M, Kimi K3 requires 1673.7 GB while Kimi K3.1 Pro needs 1673.8 GB, making Kimi K3 the easier model to run on your local setup.
Can I run Kimi K3 or Kimi K3.1 Pro on a 16 GB GPU?
Kimi K3 does not fit (smallest quant needs 766.5 GB); Kimi K3.1 Pro does not fit (smallest quant needs 1225.5 GB).
Which has a longer context window, Kimi K3 or Kimi K3.1 Pro?
Kimi K3 supports 1,048,576 tokens and Kimi K3.1 Pro supports 1,048,576 tokens.